Embed Size (px)
Citation preview
5
The Lifecycle of Derivatives Contracts
This chapter is co-authored by Michael Penick, Senior Economist
at the US Commodity Futures Trading Commission’s Office of the
Chief Economist, who provided data and feedback throughout. THIS
CHAPTER DOES NOT REFLECT THE OPINIONS OF THE
CFTC.
In chapter 4, I suggested that the natural home for teleconnection
risk is on a derivatives exchange. But my reasoning assumed that
ENSO markets will reach a sustainable level of liquidity. That assump-
tion begs a follow-up question: what is the probability that ENSO
markets will reach a sustainable level of liquidity? Or, more funda-
mentally: what is the probability that any new market will reach any
given level of liquidity?
To that end, this chapter provides:
• a statistical description of the lifecycle of exchange-traded deriva-
tives in the United States.
Using annual volumes for most derivatives reported to US ex-
changes since 1954, we present distributional estimates of the rate
at which derivative trading volumes rise and fall. Our results provide
the first published statistics on the full lifecycle of derivatives and il-
lustrate fundamental changes in cleared derivatives markets over the
2000s. In that decade, derivatives with low trading volumes moved to
modest volumes with increased probability. Prior to the 2000s, low
volume contracts were more likely to remain stuck at low volumes or
be delisted altogether. This additional resilience from low levels of
trading meant that the expected trading volume for a new cleared
derivative after ten years of trading actually grew between the 1990s
and 2000s. This is surprising given that many new contracts were
launched in the last decade and a historically large percentage of con-
tracts traded at low volume in any year.
150 direct climate markets
We also discuss the relative influence of exchange and product type
on volume patterns. We find that trading volumes varied more decade
to decade than from exchange to exchange or product type to product
type.
The results are presented as a Markov model with a non-stationary
transition matrix. Each row in the Markov model’s transition matrix
(the probabilities of a derivative moving from a given state of trading
to any other) consists of posterior draws from a Dirichlet process
estimated using Bayesian methods in R and JAGS. This approach to
the lifecycle of a derivative allowed us to make simple distributional
comparisons among subsamples (including significance testing) and
facilitated further simulation of new derivatives emerging over time.
Introduction and Literature Review
Silber [1981] and Carlton [1984] provided some of the first summary
statistics on the survival of new futures contracts. Their core conclu-
sions - that most new derivatives fail and that they do so soon after
their launch - remain widely cited1,2
. However, since those articles, 1G. Gorton and K.G. Rouwenhorst.
Facts and fantasies about commodity
futures. Working paper, 2004
2M.W. Hung, B.H. Lin, Y.C. Huang,
and J.H. Chou. Determinants of
futures contract success: Empirical
examinations for the asian futures
markets. International Review ofEconomics & Finance, 20(3):452–458,2011
technological innovation and organizational changes at derivatives ex-
changes have altered the economics of derivatives trading in ways that
may have also upended long standing patterns in product lifecycles3.
3M. Gorham and N. Singh. Electronic
Exchanges: The Global Transfor-mation From Pits to Bits. Elsevier
Science, 2009
Recently, Gorham and Kundu [2012] used a large dataset from the
Futures Industry Association (FIA) to demonstrate a steep increase in
the rate at which new futures contracts are launched.They also provide
point estimates for multiple metrics for the success of new futures
contracts. Here we extend the work in Gorham and Kundu [2012],
providing distributional estimates of contracts’ movement between
states of annual trading volume using a dataset that includes cleared
derivatives and options as well as many historical contracts that are
absent from most electronic databases.
Derivatives reform and lifecycle statistics
Basic statistics on the lifecycle of derivatives are particularly valuable
now because ongoing policy debates on derivatives regulation in the
US and Europe have hinged on projections of how new regulations will
impact liquidity and trading patterns. Better baseline statistics of the
lifecycle of derivatives, particularly statistics that take into account
recent shifts in the dynamics of exchange reported derivatives trading,
can inform that debate.
Title VII of the Dodd-Frank reforms focuses on swaps markets4,5
, 4Swaps trades have generally been
negotiated bilaterally, often over the
phone through or with large swapdealers, rather than via the central
limit order book system used by
exchange-traded derivatives. This
distinction, between markets using
bilateral negotiation and those using
central order books, has important
implications for how information
spreads among market participants
and how counter-party risk is man-
aged.
5The distinction between swaps and
futures is often murky. For example,
some swaps trades are negotiated
bilaterally and then converted into
futures trades on markets such as
the CME Group’s ClearPort. Those
trades are reported to exchanges
and are consequently included in the
dataset used in this article. The CME
and ICE, the two largest US futures
exchanges, have recently announced
plans to convert many of their most
popular swaps markets into futures
markets with physical delivery of
swaps contracts at settlement (i.e.
futures trades that become swaps),
providing yet another hybrid model.
the hitherto unregulated derivatives markets that, since the first pub-
licly disclosed swaps trade in 1981, had grown to a notional outstand-
the lifecycle of derivatives contracts 151
ing value of USD 639 trillion by June 2012. (By contrast, options and
futures had a combined notional outstanding value of USD 60 tril-
lion6.) Title VII mandates that swaps markets adopt practices related 6 Bank of International Settlements.
Statistical release: OTC derivatives
statistics at end-june 2012, June 2012.
URL http://www.bis.org/publ/
{OTC}_hy1211.pdf
to many critical market functions (such as information dissemination,
counter-party risk, and margining) comparable to those of exchange-
traded futures and options.
This regulatory change suggests that the coming years will see con-
vergence between previously unregulated swaps markets and standard
exchange-traded derivatives markets. This convergence, in turn, raises
both normative questions (How desirable is the move toward increased
clearing, public disclosure of pricing information, and greater stan-
dardization of margins?) and positive questions (What will the likely
costs or regulation be in terms of trading volume?) that would benefit
from reliable statistical descriptions of the lifecycle of derivatives.
The relative scarcity of basic statistics on the lifecycle of derivatives
has already introduced confusion into the policy debate surrounding
Title VII. In one prominent example, the International Swaps and
Derivatives Association (ISDA) released a position paper on regula-
tions mandating price transparency and clearing in swaps markets
comparable to that in exchange-traded derivatives markets in late
20117. The paper highlights previous research showing high rates of 7 ISDA Research Staff and NERA
Economic Consulting. Costs and
benefits of mandatory electronic
execution requirements for interest
rate products. Discussion Paper 2,
International Swaps and Derivatives
Association, November 2011
failure among exchange-traded derivatives. Assuming a connection
between those failure rates and exchanges’ price transparency and
clearing, the paper goes on to argue that swaps contracts subject to
proposed regulations would subsequently lose their liquidity and begin
to fail. That suggestion is misleading. First, it ignore the comparable
failure rates for bilateral swaps, which are difficult to quantify. Second,
it relies on the assumption tested here - that derivatives continue to
fail at the rates documented decades ago. Our results suggest that
assumption is not robust to recent changes in the underlying structure
of cleared derivatives markets.
In addition to providing common ground for policy debates, we
hope that the following analysis will inform the decisions of derivatives
innovators. In general, contracts are showing greater flexibility, moving
up from low levels of annual trading. This may have implications for
how exchanges allocate their limited budgets for marketing and educa-
tion. Contracts previously considered too uneven in their year-to-year
trading to succeed may indeed have substantial growth potential given
proper marketing and educational support.
Data
Our analysis is based on annual volume figures for US exchange-traded
derivatives (primarily futures, options, and cleared swaps). These fig-
152 direct climate markets
ures are/were freely available to the public through trade publications,
directly from exchanges, in newspapers, and from the website of the
Commodity Futures Trading Commission (CFTC). For ease of access,
we used:
• An electronic database maintained by the CFTC aggregating basic,
market-level daily trade data (such as volume and open interest)
regulated futures and options exchanges, called designated contract
markets (DCMs), This dataset covers all recent trading volumes
reported to US exchanges of futures, options and swaps, cleared
pursuant to DCM rules. Most contracts in that database have
volume figures dating back to the early 1980s.
• We supplemented this basic dataset by adding in futures trading
figures compiled by hand from historical publications released by
derivatives exchanges. The resulting dataset includes many short-
lived contracts listed on now-defunct exchanges that are unlikely to
appear in most electronic databases of trading statistics.
The merger of these resources may represent the most comprehen-
sive dataset on derivatives trading volume to date.
Markov model for the lifecycle of derivatives
We present our primary results in the form of a Markov model. That
model begins by imagining that a derivative contract moving be-
tween discrete states (x) of trading volume at discrete times (t as inxt)according to a discrete-time Markov chain, defined generally as in
equation 5.1.
P(Xt = xt|Xt−1 = xt−1, Xt−2 = xt−2, . . . , X0 = x0) = P(Xt = xt|Xt−1 = xt−1)
(5.1)
In the context of derivatives, the left side of equation 5.1 can be
restated as in equation 5.2.
P(Volume levelyear t+1|Volume levelyear t) (5.2)
Contract are assigned a martix (P as in equation 5.3) that describes
the probability of moving to any of a set of discrete states (time j)of annual trading volume in the following year given their state of
trading volume today (time i). This is the transition matrix commonly
used to describe a Markov process8. 8 Scott Page. Model thinking: Markovmodels. Video lectures, April 2012.URL http://www.youtube.com/watch?
v=0FumwlqGRa8
the lifecycle of derivatives contracts 153
P =
p1,1 p1,2 . . . p1,j . . .p2,1 p2,2 . . . p2,j . . ....
.... . .
.... . .
pi,1 pi,2 . . . pi,j . . ....
.... . .
.... . .
(5.3)
Volume level for any given contract-year is equivalent to the com-
mon logarithm of the annual trading, rounded down to the nearest
integer. (For example, annual trading of 10, 500 is assigned a volume
level that groups it with all contract-years with volume ≥ 10, 000 and
< 100, 000.) We assigned a special level for annual trading of 0.For ease of estimation we work with the rows of the transition
matrix P which we denote as θ. Those rows sum to 1, so, assuming
that row entries are randomly distributed, each row can be assigned a
Dirichlet distribution, commonly used for the probability of ending in
an exhaustive set of categorical states. That assignment is defined in
equation 5.4.
Volume levelyear t+1|Volume levelyear t ∼ Categorical(θ)
θ ∼ Dirichlet(xvol level 0, xvol level 1, . . . , xvol level 108)
(5.4)
We modeled these transition probabilities via Bayesian Gibbs sam-
pling through R and the Bayesian statistical package JAGS9. (We 9 M. Plummer. JAGS: A ProgramFor Analysis of Bayesian GraphicalModels Using Gibbs Sampling, 2003
used the “rjags” package10.) These methods treat the underlying prob-
10 Martyn Plummer. rjags: Bayesiangraphical models using MCMC, 2013.URL http://CRAN.R-project.org/
package=rjags. R package version3-10
abilities of moving between states of trading volume as randomly
distributed parameters, as in equation 5.4.
After estimation, we combine the vectors θ to reconstruct the tran-
sition matrix for a Markov model P. As with any Markov model, we
can multiply a vector, π0 describing the probability that a new deriva-
tive will start in any given state (at time 0) by the transition matrix
to produce a vector of probabilities that a new market will be in any
state over an arbitrary number of periods (k) as in equation 5.5.
π0Pk = πk. (5.5)
We can multiply the vector πk by yet another vector of annual
trading volumes corresponding to each possible state to get an approx-
imation of the expected trading volume in that arbitrary year. We
present all our expected trading volumes at a ten year horizon (setting
k = 10), but the Markov model is flexible in this regard.
Note that we do not assume that the transition matrix P is station-
ary across time. We measure transition matrices for various contract
groupings including decades, product categories, and exchanges to test
whether they are distinct. Given that we do not assume stationarity
154 direct climate markets
our expected value estimates do not describe an equilibrium, only the
general direction of the market.
Prior probabilities on moving between states of annual volume
Our model presumes that the data on the volume level next year
(Volume levelyear t+1) is segregated by the volume this year (Volume levelyear t)
and we assigned each of those subsets prior probabilities (corre-
sponding to parameter x in equation 5.4) of moving to any volume
level in the next year. Those priors came from an informal survey of
economists at the CFTC.
That survey found beliefs corresponding roughly to:
• Pr(Volume levelyear t+1 = Volume levelyear t−1) = 0.16
• Pr(Volume levelyear t+1 = Volume levelyear t) = 0.63
• Pr(Volume levelyear t+1 = Volume levelyear t+1) = 0.14
The probability of a contract jumping more than one order of
magnitude up or down was assigned a value of 0.01. In edge cases
(Volume levelyear t = 0 and Volume levelyear t = 108) where a move
up or down would take the contract below annual trading of 0 or to
annual trading ≥ 109, we combined the probabilities of moving up or
down with the probability of remaining in the same state. Table 4 at
the end of this article shows the full matrix of transition probability
priors.
We chose to assign informative priors on transition probabilities
because flat priors (equal weighting to the probability of a transition
to any state) unfairly biased the estimation, giving exchanges or prod-
uct subgroups with few observations a relatively high probability of
jumping to extraordinary levels of trading.
Derivatives volumes over time
Concentration of trading volume over time
Figures 5.1 and 5.2 display the empirical cumulative distribution func-
tion (ECDF) of annual trading volumes by contract for every year in
the sample. In each figure, individual lines represent the ECDFs for
a single year, with lines approaching a right angle showing greater
concentration of trading volume in a few contracts. Figure 5.1 clearly
shows that most contracts trade at low volumes in any given year,
with roughly 80 percent of contracts showing little or no volume in
any given year since 1954.0.00
0.25
0.50
0.75
1.00
0e+00 2e+08 4e+08 6e+08Annual volume
Empi
rical
CD
F
Figure 5.1: Empirical cumulativedistribution function of annual tradingvolumes by contract
However, figure 5.1 obscures substantial variation in the concen-
tration of volume over time. Figure 5.2 zooms in on the same annual
the lifecycle of derivatives contracts 155
ECDFs displayed in figure 5.1. The ECDF for each year is colored
chronologically, with the lines representing the oldest years in the sam-
ple in red and the most recent years in purple. Each panel of figure 5.1
shows the same ECDFs, but the years in a specific decade are high-
lighted (in black) to give a sense of how concentration has varied from
decade to decade.
In this graphic we see clear patterns in concentration over time.
Markets grew steadily less concentrated between the 1950s and 1990s
(perhaps with some retrenchment between the 1980s and 1990s),
shown by flattening ECDFs for each succeeding decade. That trend
reversed sharply in the 2000s, with the annual ECDFs approaching a
right angle. In the 1980s the range of 15,000 to 30,000 roughly marked
the 50th percentile for annual trading volumes, with half of the listed
contracts trading above that range and half below. By the 2000s that
range had fallen to between 300 and 8,000.
Figure 5.2 itself highlights one likely cause of this shift - the ex-
plosion of innovation during the 2000s. The ECDFs for the 2000s
are appreciably smoother than those of previous decades, with 2011
looking almost like a continuous function. This smoothness is due to
the inclusion of additional contracts. Figure 5.3 directly displays the
number of contracts with annual reported volume (which is allowed
be zero) in the sample by year. It shows the same explosive trend in
innovation discussed in Gorham and Kundu [2012], with over 3000
derivatives contracts reporting annual volume in 2011.
156 direct climate markets
0.50.60.70.80.91.0
0.50.60.70.80.91.0
0.50.60.70.80.91.0
0.50.60.70.80.91.0
0.50.60.70.80.91.0
0.50.60.70.80.91.0
0.50.60.70.80.91.0
1950's1960's
1970's1980's
1990's2000's
2011
0e+00 1e+06 2e+06 3e+06 4e+06 5e+06Annual volume
Empi
rical
CD
F
196019701980199020002010
Year
Figure 5.2: Empirical cumulative dis-
tribution function (ECDF) of annual
trading volumes by contract with
scale adjusted to distinguish between
decades - Each line represents the
ECDF for a different year. Each of
the stacked panels highlights the
years in a particular decade in black.
Note that an ECDF approaching a
right angle represents a year in which
volume was concentrated in a few
contracts. Hence, with some excep-
tions in the 1990s the market as a
whole becomes less concentrated, until
the 2000s when it abruptly becomes
highly concentrated.
the lifecycle of derivatives contracts 157
0
1000
2000
3000
1960 1970 1980 1990 2000 2010Year
Der
ivat
ives
in s
ampl
e
Figure 5.3: Number of contracts insample by year
Probability of individual contracts moving to different levels of trad-
ing by decade
Figures 5.4 and 5.5 give the probabilities of individual contracts mov-
ing between volume levels in a given year t (indicated by the row of
estimates) and volume levels in year t+1 (indicated by the column of
estimates). These probabilities, estimated separately for each decade
in the sample via equation 5.4, combine to form the transition matrix
for a Markov model of a contract emerging over time.
The parameter estimates indicate that there is substantial inertia
across every decade keeping contracts with a given level of trading vol-
ume at that same volume in the following year. In virtually all decades
in the sample, contracts trading at or above 1,000 in annual volume
were more likely to remain at their trading volume level than to move
up or down. This dynamic is particularly strong at higher levels of
trading. In most decades where relevant observations were available,
contracts with annual volume of one million or above remained in
that range the following year with probabilities between ∼80 and ∼90
percent (see lower right-hand corner of figure 5.5).
158 direct climate markets
vol 0, year t+1 vol 1's, t+1 vol 10's, t+1 vol 100's, t+1 vol 1000's, t+1 vol 10^4's, t+1 vol 10^5's, t+1 vol 10^6's, t+1 vol 10^7's, t+1 vol 10^8's, t+1
!
!
!
!
!
!
!
0.53
0.6
0.89
0.93
0.82
0.7
0.46
!
!
!
!
!
!
!
0.59
0.46
0.75
0.45
0.61
0.48
0.58
!
!
!
!
!
!
!
0.3
0.55
0.62
0.58
0.51
0.32
0.47
!
!
!
!
!
!
!
0.02
0.13
0.25
0.29
0.23
0.16
0.26
!
!
!
!
!
!
!
0
0.03
0.1
0.15
0.14
0.1
0.11
!
!
!
!
!
!
!
0.1
0.13
0.02
0
0.01
0.03
0.17
!
!
!
!
!
!
!
0.1
0.28
0.08
0.09
0.15
0.14
0.18
!
!
!
!
!
!
!
0.16
0.06
0.15
0.05
0.11
0.08
0.15
!
!
!
!
!
!
!
0.04
0.05
0.06
0.11
0.07
0.02
0.09
!
!
!
!
!
!
!
0.01
0
0.04
0.02
0.02
0.01
0.02
!
!
!
!
!
!
!
0.08
0.13
0
0.03
0.04
0.13
0.15
!
!
!
!
!
!
!
0.21
0.09
0.14
0.03
0.02
0.2
0.15
!
!
!
!
!
!
!
0.21
0.29
0.2
0.12
0.16
0.22
0.22
!
!
!
!
!
!
!
0.27
0.24
0.17
0.12
0.19
0.13
0.23
!
!
!
!
!
!
!
0
0.05
0.06
0.14
0.04
0.04
0.07
!
!
!
!
!
!
!
0.2
0.06
0.05
0.01
0.06
0.08
0.14
!
!
!
!
!
!
!
0.05
0.09
0
0.1
0.08
0.12
0.06
!
!
!
!
!
!
!
0.31
0.08
0
0.05
0.13
0.27
0.11
!
!
!
!
!
!
!
0.56
0.49
0.2
0.09
0.33
0.43
0.31
!
!
!
!
!
!
!
0.17
0.25
0.15
0.08
0.21
0.18
0.25
!
!
!
!
!
!
!
0
0.04
0
0.01
0.02
0.04
0.04
!
!
!
!
!
!
!
0
0
0.02
0.14
0.08
0.04
0.02
!
!
!
!
!
!
!
0
0.01
0.01
0.05
0.03
0.09
0.04
!
!
!
!
!
!
!
0.09
0.06
0.29
0.15
0.12
0.22
0.1
!
!
!
!
!
!
!
0.7
0.53
0.5
0.29
0.4
0.48
0.45
!
!
!
!
!
!
!
0
0.02
0.02
0
0.03
0
0.03
!
!
!
!
!
!
!
0
0.04
0
0.06
0.02
0
0
!
!
!
!
!
!
!
0
0
0
0.1
0.01
0.02
0.01
!
!
!
!
!
!
!
0
0.01
0.02
0.13
0.03
0.05
0.01
!
!
!
!
!
!
!
0.11
0.11
0.12
0.14
0.17
0.18
0.1
!
!
!
!
!
!
!
0
0
0
0.01
0.01
0
0.01
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0.02
0
0
!
!
!
!
!
!
!
0
0
0
0.05
0.02
0
0
!
!
!
!
!
!
!
0
0.02
0.03
0.08
0.01
0.01
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0.06
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0.01
0
0
!
!
!
!
!
!
!
0
0
0
0.03
0
0
0
!
!
!
!
!
!
!
0
0
0
0.08
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
vol 0, year tvol 1's, t
vol 10's, tvol 100's, t
vol 1000's, t
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
Pr(year−on−year move)
Dec
ade
Figure 5.4: Transition matrix forMarkov model of derivatives contractmoving between states of annualtrading volume by decade - rowrepresents state in year t, columnrepresents state in year t + 1, medianestimate indicated by dot, 95 percentprobability interval indicated by line- part 1: transitions given annualvolumes ≥ 0 and < 10, 000
the lifecycle of derivatives contracts 159
vol 0, year t+1 vol 1's, t+1 vol 10's, t+1 vol 100's, t+1 vol 1000's, t+1 vol 10^4's, t+1 vol 10^5's, t+1 vol 10^6's, t+1 vol 10^7's, t+1 vol 10^8's, t+1
!
!
!
!
!
!
!
0.01
0.01
0.03
0.05
0.05
0.04
0.05
!
!
!
!
!
!
!
0
0
0
0.01
0.01
0.01
0.01
!
!
!
!
!
!
!
0
0
0
0.01
0.01
0.01
0
!
!
!
!
!
!
!
0
0
0
0
0.03
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0.01
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0.01
0.02
0.01
0.01
0.01
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0.01
0.04
0.04
0.02
0.04
0.04
!
!
!
!
!
!
!
0
0
0
0.01
0.01
0
0
!
!
!
!
!
!
!
0
0
0
0.01
0
0
0
!
!
!
!
!
!
!
0
0
0
0.01
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0.09
0.13
0.09
0.13
0.13
0.21
0.25
!
!
!
!
!
!
!
0
0
0.01
0.04
0.02
0.02
0.01
!
!
!
!
!
!
!
0
0
0
0.02
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0.85
0.7
0.58
0.61
0.71
0.61
0.56
!
!
!
!
!
!
!
0.12
0.11
0.05
0.07
0.09
0.17
0.22
!
!
!
!
!
!
!
0
0
0
0.02
0
0.01
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0.05
0.15
0.24
0.1
0.07
0.09
0.1
!
!
!
!
!
!
!
0.84
0.83
0.84
0.7
0.81
0.72
0.67
!
!
!
!
!
!
!
0.77
0.15
0.04
0.08
0.05
0.08
0.19
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0.04
0
0
0
!
!
!
!
!
!
!
0.03
0.06
0.1
0.14
0.06
0.07
0.09
!
!
!
!
!
!
!
0.15
0.83
0.94
0.8
0.91
0.81
0.78
!
!
!
!
!
!
!
0.02
0.02
0.02
0.17
0.02
0.02
0.07
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0.02
0.05
0.02
0.09
0.02
!
!
!
!
!
!
!
0.74
0.74
0.74
0.81
0.91
0.92
0.93
!
!
!
!
!
!
!
0.02
0.02
0.02
0.02
0.84
0.08
0.06
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0
0
0
0
0
0
0
!
!
!
!
!
!
!
0.01
0.01
0.01
0
0.04
0.04
0
!
!
!
!
!
!
!
0.92
0.92
0.92
0.92
0.14
0.92
0.94
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
1950
1960
1970
1980
1990
2000
2010
vol 10^4's, tvol 10^5's, t
vol 10^6's, tvol 10^7's, t
vol 10^8's, t
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
Pr(year−on−year move)
Dec
ade
Figure 5.5: Transition matrix forMarkov model of derivatives contractmoving between states of annualtrading volume by decade - rowrepresents state in year t, columnrepresents state in year t + 1, medianestimate indicated by dot, 95 percentprobability interval indicated by line- part 2: transitions given annualvolumes ≥ 10, 000
160 direct climate markets
vol 0, year t+1
!
!
!
!
!
!
!
0.53
0.6
0.89
0.93
0.82
0.7
0.46
1950
1960
1970
1980
1990
2000
2010
vol 0, year t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Dec
ade
Figure 5.6: Probability of remainingat annual volume of zero from year toyear by decade
We also see substantial historical evidence of inertia at very low
levels of trading. From 1970 until 2000, the median probability that
a contract with trading volume of zero would remain at zero the next
year, ranged between 80 and 95 percent (see upper left-hand corner of
figure 5.4).
The transition matrix begins to depart from the prevailing story in
Silber [1981] and Carlton [1984] when you look at contracts at lower
levels of trading in the 2000s. (See the top rows of figure 5.4.) The
inertia for those contracts is lower than in previous decades, with
the median probability of a contract at an annual volume of zero re-
maining at zero falling to 70 percent (figure 5.6). While zero volume
contracts remained unlikely in absolute terms to rise to higher volume
levels, the 95 percent probability interval for the transition probability
for the 2000s does not overlap with those for recent decades, meaning
that the difference holds with high probability.
vol 1's, t+1
!
!
!
!
!
!
!
0.1
0.13
0.02
0
0.01
0.03
0.17
1950
1960
1970
1980
1990
2000
2010
vol 0, year t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Dec
ade
vol 10's, t+1
!
!
!
!
!
!
!
0.21
0.09
0.14
0.03
0.02
0.2
0.15
1950
1960
1970
1980
1990
2000
2010
vol 1's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Dec
ade
Figure 5.7: Probability of transitionfrom annual volume of 0 to annualvolume in the single digits (left) andfrom annual volume in the singledigits to annual volume ≥ 10,< 100(right)
During the decade of the 2000s, contracts were substantially more
likely to jump from an annual trading volume of 0 to trading volumes
between 10 and 1000 than in previous decades. (See top row of figure
5.4 and 5.7.) Combined with the apparent trend toward maintaining
rather than delisting contracts, this suggests that there was less path-
dependence for trading volumes in the 2000s. While more contracts
traded at a volume of 0 in any given year (see figure 5.2), contracts
were substantially more likely to jump up from such low trading vol-
umes in the 2000s.
Having reached annual trading volumes in the 10s or 100s (see 5.8),
contracts in the decade of the 2000s were again substantially more
likely to continue increasing their trading volume in the 2000s than in
the 1980s or 1990s. Only after reaching trading volumes in the 1000s
(figure 5.9) did the probability of an individual contract progressing
to higher levels of annual trading volume fall roughly back within the
same range as those from previous decades. In the 2000s, contracts
generally moved up to annual trading in the thousands with an ease
not seen in previous decades.
the lifecycle of derivatives contracts 161
vol 100's, t+1
!
!
!
!
!
!
!
0.31
0.08
0
0.05
0.13
0.27
0.11
1950
1960
1970
1980
1990
2000
2010
vol 10's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Dec
ade
vol 1000's, t+1
!
!
!
!
!
!
!
0.09
0.06
0.29
0.15
0.12
0.22
0.1
1950
1960
1970
1980
1990
2000
2010
vol 100's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Dec
ade
Figure 5.8: Probability of transitionfrom annual volume in the hundredsto annual volume in the thousands(left) and from annual volume in thethousands to annual volume tens ofthousands (right)
vol 10^4's, t+1
!
!
!
!
!
!
!
0.11
0.11
0.12
0.14
0.17
0.18
0.1
1950
1960
1970
1980
1990
2000
2010
vol 1000's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Dec
ade
Figure 5.9: Probability of transitionfrom annual volume in the tens ofthousands to annual volume in thehundreds of thousands
Contracts trading in the tens of thousands were 8 percent more
likely to fall back to lower levels of annual volumes in the 2000s than
in previous decades, a difference that holds with high probability.
This indicates that some of the flexibility gained for contracts at lower
levels of trading may have come at the expense of contracts at mid
to high levels of trading. (However, as we see in figure 5.12, discussed
below, that retrenchment from trading in the tens of thousands was
not enough, on balance, to lower the prospects of a new contract over
the course of ten years.)
volume level vol 0’s, t+1 1 10 100 1000 104 105 106 107 108
vol 0’s, t 0.66 0.08 0.11 0.09 0.03 0.02 0.00 0.00 0.00 0.001 0.56 0.16 0.16 0.08 0.03 0.01 0.00 0.00 0.00 0.0010 0.42 0.12 0.21 0.18 0.06 0.02 0.00 0.00 0.00 0.00100 0.20 0.05 0.17 0.38 0.17 0.03 0.00 0.00 0.00 0.001000 0.10 0.01 0.05 0.20 0.47 0.15 0.01 0.00 0.00 0.00104 0.04 0.00 0.01 0.03 0.19 0.62 0.10 0.00 0.00 0.00105 0.01 0.00 0.00 0.00 0.02 0.14 0.74 0.08 0.00 0.00106 0.01 0.00 0.00 0.00 0.01 0.01 0.08 0.84 0.04 0.00107 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.05 0.90 0.03108 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.13 0.87
Table 5.1: Median estimates of tran-sition matrix between volume stateson full sample - with annual tradingvolume state in year t denoted byrow, trading volume state in year tdenoted by column
Annual trading in the 10,000s appears to represent an important
milestone for contracts across the sample. Having reached this level
of trading, the likelihood of outright collapse (annual trading volume
falling to 0 in the next year) fell to very low levels and was largely in-
distinguishable across the decades (figure 5.10). Table 5.1 presents the
median estimates of transition probabilities estimated across the full
sample (i.e. aggregating across decades). They show clearly that hav-
ing reached annual trading of in the 10,000s, a full collapse becomes
relatively unlikely (4 percent). In fact, for contracts that achieve an-
nual trading in the 10,000s, the probability of falling more than one
162 direct climate markets
volume level is below 10 percent. (See the sixth row of table 5.1.)
Note that these full sample estimates are biased toward recent decades
because the sample contains more observations from recent decades.
vol 0, year t+1
!
!
!
!
!
!
!
0.01
0.01
0.03
0.05
0.05
0.04
0.05
1950
1960
1970
1980
1990
2000
2010
vol 10^4's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Dec
ade
Figure 5.10: Probability of transitionfrom annual volume in the tens ofthousands to annual volume of 0
As suggested above, one hypothesis regarding the recent shift in
derivatives lifecycles is that the additional flexibility that low volume
contracts enjoyed in the 2000s came directly at the expense of mid-
range to higher volume contracts. In volatile markets, hedgers might
be choosing niche contracts with lower basis risk over more liquid
cross-hedges. What would that mean for the overall outlook for life-
time trading of derivatives? We test this by looking at the expected
trading volume of a new derivative over the course of ten years.
Combining draws from the transition matrix in figures 5.4 and 5.5
with draws from a vector representing the probability of a contract
starting in each of the available states of annual trading volume (esti-
mated using the same basic model presented in equation 5.4) we can
get the probability that a new contract will be in any given state of
volume after ten years of trading. Those values are displayed in fig-
ure 5.11. Figure 5.11 makes clear the resilience of contracts trading
at low levels in the 2000s. Only 32 percent of contract that debuted
with zero volume were still trading at zero volume after ten years in
the simulation representing the 2000s. Those probabilities were 46,
48, and 52 percent in the 1990s, 1980s and 1970s respectively (See the
first column of boxes in figure 5.11).11 Instead of languishing, con- 11 Note these simulated values simplydescribe the dynamics of the tran-sition matrices when compounded.They ignore delisting. If we accountedfor delisting, a practice that was morecommon in previous decades, theprobabilities of failure would likely behigher for those decades.
tracts simulated from the 2000s were more likely to migrate over ten
years to moderate levels of trading. (See the columns of boxes in figure
5.11 corresponding to annual trading volume between 100 and 10,000.)
Those same contracts were, however, less likely to reach the highest
levels of trading (≥ 100, 000) than contracts from other decades. The
1980s appears to be the best decade for such blockbuster contracts, as
suggested in Gorham and Kundu [2012].
Simply comparing the raw probabilities of reaching various levels of
volume after ten years, it is difficult to discern which decade provided
a better overall environment for new contracts. To make that compar-
ison, we normalize the probabilities in figure 5.11 by the lower bound
of each trading range (i.e. multiplying the probability of being in the
trading state ≥ 100 and < 1, 000 by 100). This give an approxima-
tion of the expected trading volume of a new contract after ten years,
displayed in figure 5.12. Based on that graph, we can conclude:
• The expected trading volume after 10 years for a contract has var-
ied substantially from decade to decade;
• There is no clear trend that emerges from these variations over
time;
• The expected trading volume at year ten for a contract in the 2000s
the lifecycle of derivatives contracts 163
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!
!!!
!
!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!
!
!!!!!!!!!!!!!!!!!!!
!
!!!!!
!!
!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!
!
!!!!!
!!
!
!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!
!!
!!!!!!!!!!!!!!
!
!!
!
!!!!!!!!!!!!!!!
!
!!!!!
!
!
!!
!!!
!
!!
!!!!!!
!
!!
!
!!!!!!
!
!!!!!
!!
!!!!
!
!!!!!!!!!!!!!!!
!
!
!
!
!
!!!
!
!!!!!!!!
!
!!!!!!!!!
!
!!
!!!
!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
0.150.05 0.11
0.210.14
0.220.1
0.01 0 0
!!
!!!!
!!
!!!!
!!!!!!
!!!
!!!!!!!!!
!
!!!!
!
!!!!!!!
!!
!!
!
!
!!
!!!!
!
!!
!!!
!
!
!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!
!
!!
!
!!!!!!!!!!!!!!
!
!!!!!
!
!!!!!!!!!!
!
!!!!!!!
!
!!
!
!!!!!!!!!!!!!
!
!!!!!!!!
!
!
!
!!!!!!!!! !
!
!!!!!!!!!!!!!!!!!!!!!!!!
!!
!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!
!
!
!
!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!
!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!
!
!! !!!!!!!
!
!
!!
!!
!
!!
!
!!!!!!!!
!
!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!
!
!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
0.33
0.08 0.13 0.13 0.09 0.11 0.10.02 0 0
!!!!!!!!!!!!!!
!!
!!!
!!
!!!!
!
!!
!
!!!!!
!
!
!
!!!
!
!!!!
!
!!
!!
!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!
!!!!!!!!!!
!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!
!
!!!
!
!!
!
!!!!!!
!
!!!!!
!
!
!
!!!!!!!!!!!!!!!
!
!
!
!!!!!!!!!!!!!!!!
!
!
!
!!!!!!
!
!!!!!!
!
!!!!!!
!
!!!!!!!!!
!!
!!!!
!
!!!
!
!!!!!!!!!!!!!!!!
!
!!!!!!!!
!
!!!!!!!!!
!
!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!
!
!!
!
!!!!
!
!
!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
0.52
0.02 0.02 0.05 0.05 0.070.15 0.1
0 0
!
!!
!
!
!
!!
!
!!
!
!!
!
!!
!!!
!
!
!!
!
!!!
!
!!!!
!
!
!
!!!
!
!
!
!
!!
!
!!!
!!!!!!
!!
!
!
!!!
!
!!!
!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!
!
!
!
!!!!
!
!!!!!!!!!!!!!!!!
!
!!!!!!
!
!!!!
!
!!!!!!!
!
!!!!!!!!!!
!!!!!!!
!
!!!!!!!
!
!
!
!!
!
!!
!!!
!
!
!!!!!!
!!
!!!!!!
!
!!!
!
!!!!
!!
!
!!
!
!
!!!
!
!
!!
!!
!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
0.48
0.01 0.03 0.02 0.04 0.08 0.12 0.170.04 0
!!
!
!!!!
!!
!
!!!
!!
!
!!!!!
!
!!!
!
!
!
!!!!!!!
!!
!!
!!!!
!
!!
!
!
!!!!
!!
!!!!!!
!
!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!
!
!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !
!
!
!!!!!!!
!!
!!!!!!!!
!
!!!!
!
!!!!!!
!!
!!!
!
!!!!!!!!
!
!!!!!!!!!!
!
!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
0.46
0.02 0.05 0.09 0.080.15 0.1
0.04 0.01 0
!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
0.32
0.030.1
0.16 0.17 0.130.05 0.02 0.01 0
!
!!
!
!!!
!
!
!!
!!!
!!!!
!!
!!
!
!!
!!!
!
!
!
!
!
!
!!!!
!!
!
!!
!
!!!
!
!
!!!
!
!
!!!
!!!!
!!
!
!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
0.35
0.12 0.14 0.15 0.1 0.07 0.04 0.02 0 0
0.00.20.40.6
0.00.20.40.6
0.00.20.40.6
0.00.20.40.6
0.00.20.40.6
0.00.20.40.6
0.00.20.40.6
19501960
19701980
19902000
2010
vol 0, year 10vol 1's, year 10vol 10's, year 10vol 100's, year 10vol 1000's, year 10vol 10^4's, year 10vol 10^5's, year 10vol 10^6's, year 10vol 10^7's, year 10vol 10^8's, year 10
Volume level
Pr(B
eing
in a
give
n st
ate
afte
r 10
year
s
Figure 5.11: Box and whiskers plotof probability of a new contractbeing at different levels of tradingafter 10 years by decade - mediansimulated probability marked intext, upper and lower hinges of thebox plot correspond to the first andthird quartiles (the 25th and 75thpercentiles)
164 direct climate markets
was firmly in the middle of the historical range - the 2000s were
lower than the 1980s, higher than the 1990s, and all three decades
showed substantial overlap with the earlier decades in the sample;
• In the 2000s, low volume contracts tended to rise to modest levels
of trading, balancing any fall in the probability of reaching the
highest trading levels.
!
!
!
!
!
!
!
1960
1980
2000
1e+05 1e+06Expected volume after 10 years
Dec
ade
Figure 5.12: Expected trading volumeover ten years by decade
While a larger percentage of contracts were at low volumes in the
2000s than in previous decades (figure 5.1), individual contracts were
considerably more likely to jump up from very low volumes to mod-
erate volumes (figure 5.7). The net effect of these trends set the ex-
pected volume of contracts at year ten well within the historical range
of earlier decades (figure 5.12). This is remarkable given the explo-
sion in the number of contracts launched (figure 5.3). It suggests that
the marginal value of an innovative contract (approximated by its ex-
pected trading volume at year ten) did not fall in the 2000s, despite
exponentially higher rates of innovation than in past decades.
This shift is consistent with the hypothesis that electronic trad-
ing made trading activity more mobile across derivatives markets and
substantially cut the costs of launching and sustaining a derivatives
contract. But changes went above and beyond the introduction of elec-
tronic trading on US and European exchanges in the 2000s, making
it difficult to identify the causes of product lifecycle shifts in aggre-
gate statistics. For example, many of the new contracts launched in
the 2000s (and included in this sample) are bilaterally-negotiated,
the lifecycle of derivatives contracts 165
but centrally-cleared swaps. In the wake of Enron’s collapse, which
threatened energy firms with counter-party defaults on their swaps
trades, exchanges launched popular new facilities devoted to these
cleared-swaps, including the CME’s ClearPort. While those contracts
benefited from a suite of tools associated with electronic trading, they
were not subject to electronic trading in the narrow sense of actually
having buy and sell orders matched on an electronic platform.
To isolate the influence of electronic trading, we look at contracts
trading on the New York Mercantile Exchange (NYMEX), where
electronic trading was introduced suddenly. The NYMEX does not
offer an ideal natural experiment. Its trading patterns were likely
influenced by the shift toward cleared swaps throughout the 2000s.
However, the abruptness of the exchange’s switch to electronic trading
does offer some scope for teasing out the relative import of electronic
trading.
Derivatives volumes by exchange
Differences in trading volume patterns over the life of a derivatives
contract may be influenced by the exchange offering the contract.
Carlton [1984] hypothesized that economies of scale in designing and
launching a contract gave those on larger exchanges a relative advan-
tage in terms of trading volumes. Similarly, there may be network
effects stemming from an exchange’s ability to cross-margin trades.
Cuny [1993] and Holland and Fremault [1997] suggest that inno-
vative exchanges may enjoy a first-mover advantage, capturing a dis-
proportionate share of trading on those contracts that they launch.
Gorham and Kundu [2012] tests this hypothesis and finds little persis-
tent advantage. In the context of a Markov model of trading volumes,
if indeed there is a first-mover advantage, then we would expect in-
novative exchanges to distinguish themselves with higher expected
trading in year ten.
Figure 5.13 presents expected volume in year ten for contracts
on all exchanges in the sample. Contracts show greater distinction
across decades (as in figure 5.12) than across exchanges. It is possible
to distinguish individual exchanges from one another. For example,
contracts on the Chicago Board of Trade have an advantage over those
on the NYMEX in expected value terms. But no exchanges clearly
distinguish themselves from the general tendency with greater than 95
percent probability. Possible exceptions include:
• the single-stock futures traded on OneChicago which show particu-
larly low expected trading volumes over ten years
• the two registered exchanges in the IntercontinentalExchange group,
166 direct climate markets
marked ICE and ICEU in figure 5.13, which likely have higher ex-
pected trading volumes than most other exchanges. It is important
to note that these exchanges specialize in OTC markets, only a
handful of which have been reported to the CFTC as futures. Con-
sequently, some of their performance may represent selection bias.12 12 In late 2012, the Intercontinen-talExchange announced that many ofits most popular OTC contracts willbegin trading as futures.
CME acquisitions test the important of exchange to lifecycles
Recent exchange acquisitions offer the chance to test the effects of
particular exchanges on trading volumes. Gorham and Kundu [2012]
singled out the CME as the exchange with a persistent advantage over
its rivals - leading other major exchanges in mean volume in the 5th
year of trading, mean lifetime volume, and their approximations of
present value discounted fee generation. In the late 2000s, the CME
Group effectively13 took over both the New York Mercantile Exchange 13 Technically, the CME and CBOTmerged. However, the CME was thedominant firm in the merger, initiat-ing the transaction and retaining mostof the key staff positions. Olson [2010]provides an inside account of the fightbetween the CME and ICE for controlof the CBOT.
(designated in the database as NYME but commonly referred to as
the NYMEX) and the Chicago Board of Trade (CBT). After the ac-
quisitions, the exchanges’ contracts continued to be reported as before
(i.e. NYMEX contracts continued to be reported in the dataset as
NYMEX contracts).
If indeed the CME did enjoy a persistent advantage on multiple
volume metrics, then presumably the transition matrices for NYMEX
and CBOT contracts, calculated using the Markov models profiled
here, would improve following their acquisitions. These acquisitions
could also test a weaker form of that same hypothesis. If exchange
management is important to contract lifecycles, then the CBOT
and NYMEX’s contracts’ transition matrices should converge to the
CME’s, regardless of whether the CME has an advantage over other
exchanges or not.
Figures 5.14 and 5.15 for the NYMEX and Lifecycle Appendix’s fig-
ures 18 and 19 present the transition matrices for each of the merged
exchanges in the years before and after the merger.14 14 We chose to present the full tran-sition matrix for the exchange-yearcomparisons rather than the expectedvalue figures because we believe thatthe former provide more robust infer-ence. Expected value calculations aresensitive to the initial trading volumesof the contracts that happened tolaunch after the merger.
The CBT’s transition matrices (Lifecycle Appendix’s figures 18
and 19) show no consistent trends in post-merger years relative to the
earlier years in the sample. Post-merger years with strong performance
(contracts showing a high probability of advancing to a higher level
of liquidity - such as 2010, where many of the contracts previously
trading with annual volumes in the thousands advanced to the tens
of thousands) do not stand out relative to the pre-merger era. To
the extent that the CBT shows any post-merger trend, it stems from
2010, an especially volatile year for the CBT, where many contracts
advanced to trading in the tens of thousands and a particularly large
percentage fell back from annual volumes in the tens of thousands.
Unlike the transition matrix for the CBT, the NYMEX shows a
the lifecycle of derivatives contracts 167
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
ACE
BCC
C
CBT
CCX
CME
CMX
CRCE
E
ELX
EUR
EUS
GE
ICE
ICEU
ICEX
ICUS
IEPA
KCBT
MCE
MESL
MGE
Memphis
MilGX
N
NOCE
NYBOT
NYL
NYL2
NYME
NYPE
PANY
PBOT
PCE
Seattle
West Coast
1e+05 1e+07Expected volume after 10 years
Exch
ange
Figure 5.13: Expected trading volumeover ten years by exchange
168 direct climate markets
vol 0, year t+1 vol 1's, t+1 vol 10's, t+1 vol 100's, t+1 vol 1000's, t+1 vol 10^4's, t+1 vol 10^5's, t+1 vol 10^6's, t+1 vol 10^7's, t+1 vol 10^8's, t+1
!
!
!
!
!
!
!
!
!
!
0.98
0.35
0.94
0.98
0.46
0.69
0.81
0.58
0.35
0.43
!
!
!
!
!
!
!
!
!
!
0.02
0.02
0.02
0.77
0.53
0.36
0.55
0.66
0.66
0.45
!
!
!
!
!
!
!
!
!
!
0
0
0.5
0.31
0.27
0.36
0.19
0.29
0.32
0.44
!
!
!
!
!
!
!
!
!
!
0
0.21
0.11
0.06
0.19
0.29
0.3
0.18
0.3
0.25
!
!
!
!
!
!
!
!
!
!
0.04
0
0
0
0.1
0.05
0.08
0.16
0.11
0.16
!
!
!
!
!
!
!
!
!
!
0
0
0.01
0
0
0
0
0
0.07
0.05
!
!
!
!
!
!
!
!
!
!
0.73
0.73
0.74
0.15
0.07
0.15
0.39
0.13
0.03
0.21
!
!
!
!
!
!
!
!
!
!
0.01
0.01
0.01
0
0
0
0.09
0.03
0
0.04
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0.01
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0.5
0.08
0.05
0
0.23
0.08
!
!
!
!
!
!
!
!
!
!
0.01
0.01
0.01
0
0.34
0
0
0
0.18
0.29
!
!
!
!
!
!
!
!
!
!
0.25
0.25
0.25
0.07
0.25
0.05
0.22
0.35
0.26
0.34
!
!
!
!
!
!
!
!
!
!
0
0
0.13
0.11
0.1
0.07
0.03
0.09
0.12
0.11
!
!
!
!
!
!
!
!
!
!
0.04
0.06
0
0
0
0
0
0
0.01
0.01
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0.18
0.11
0.22
0.15
0.2
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0.29
0
0.07
0.07
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0.53
0.28
0.44
0.3
0.22
0.27
0.16
!
!
!
!
!
!
!
!
!
!
0.07
0.39
0.21
0.31
0.28
0.33
0.37
0.41
0.44
0.49
!
!
!
!
!
!
!
!
!
!
0
0
0.15
0.1
0.16
0.13
0.08
0.09
0.15
0.22
!
!
!
!
!
!
!
!
!
!
0
0.5
0
0
0
0
0
0.13
0.11
0.02
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0.07
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0.1
0.11
0.04
0.06
0.13
0
!
!
!
!
!
!
!
!
!
!
0.34
0
0.13
0.41
0.4
0.24
0.29
0.28
0.13
0.13
!
!
!
!
!
!
!
!
!
!
0.07
0.19
0.56
0.48
0.4
0.42
0.48
0.56
0.56
0.52
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0.03
0.03
0.14
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.5
0
0
0
0.04
0
0.09
0
0
0
!
!
!
!
!
!
!
!
!
!
0.32
0.21
0.27
0.06
0
0.04
0
0.02
0.01
0.01
!
!
!
!
!
!
!
!
!
!
0.37
0.71
0.21
0.39
0.33
0.39
0.31
0.19
0.15
0.07
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0.02
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0.5
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.13
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.42
0
0.04
0
0
0
0.04
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0.02
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
vol 0, year tvol 1's, t
vol 10's, tvol 100's, t
vol 1000's, t
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
Pr(year−on−year move)
Exch
ange−Y
ear
Figure 5.14: Transition matrix forMarkov model of derivatives con-tract moving between states ofannual trading volume by decade- NYMEX before and after CMEmerger (announced March 2008, fi-nalized September 2009) and beforeand after switch to electronic trading(September 2006) - part 1: transi-tions given annual volumes ≥ 0 and< 10, 000
the lifecycle of derivatives contracts 169
vol 0, year t+1 vol 1's, t+1 vol 10's, t+1 vol 100's, t+1 vol 1000's, t+1 vol 10^4's, t+1 vol 10^5's, t+1 vol 10^6's, t+1 vol 10^7's, t+1 vol 10^8's, t+1
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0.06
0.01
0
0.07
0.03
0.06
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0.01
0
0.01
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0.07
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0.01
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0.02
0
0
0
0.01
0.01
0.01
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0.08
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0.02
0.04
0.01
0.01
0.03
0
0.05
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0.1
0.18
0.08
0.04
0.2
0.22
0.23
0.29
!
!
!
!
!
!
!
!
!
!
0
0
0
0.02
0
0
0
0.03
0
0.04
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.25
0.69
0.52
0.71
0.74
0.73
0.59
0.61
0.67
0.55
!
!
!
!
!
!
!
!
!
!
0
0.09
0.07
0.1
0.11
0
0.29
0.41
0.31
0.38
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.73
0.3
0.36
0.05
0.07
0.2
0.17
0.05
0.05
0.04
!
!
!
!
!
!
!
!
!
!
0.69
0.9
0.72
0.73
0.88
0.97
0.62
0.53
0.63
0.56
!
!
!
!
!
!
!
!
!
!
0.02
0.01
0.01
0
0.18
0.25
0.16
0.47
0.34
0.41
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0.02
0
0
0
!
!
!
!
!
!
!
!
!
!
0.25
0
0.19
0.13
0
0.03
0.08
0.01
0.05
0
!
!
!
!
!
!
!
!
!
!
0.74
0.91
0.25
0.8
0.61
0.59
0.8
0.51
0.63
0.46
!
!
!
!
!
!
!
!
!
!
0
0
0
0.16
0.16
0
0
0.16
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.01
0
0.59
0.15
0.08
0.11
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.97
0.97
0.97
0.8
0.62
0.97
0.98
0.8
0.97
0.97
!
!
!
!
!
!
!
!
!
!
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0
0
0
0
0.16
0
0
0
0
0
!
!
!
!
!
!
!
!
!
!
0.92
0.92
0.92
0.92
0.92
0.97
0.97
0.97
0.97
0.97
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
NYME−2001NYME−2002NYME−2003NYME−2004NYME−2005NYME−2006NYME−2007NYME−2008NYME−2009NYME−2010
vol 10^4's, tvol 10^5's, t
vol 10^6's, tvol 10^7's, t
vol 10^8's, t
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
0.00
0.25
0.50
0.75
1.00
Pr(year−on−year move)
Exch
ange−Y
ear
Figure 5.15: Transition matrix forMarkov model of derivatives con-tract moving between states ofannual trading volume by decade- NYMEX before and after CMEmerger (announced March 2008, fi-nalized September 2009) and beforeand after switch to electronic trading(September 2006) - part 2: transitionsgiven annual volumes ≥ 10, 000
170 direct climate markets
clear trend in its transition probabilities. On the rows in figures 5.14
and 5.15 indicating trading volume between ≤ 10 and < 1, 000, 000(rows three through five in figure 5.14 and rows one and two in figure
5.15), a gradual pattern in volume level transitions emerges that is
strong enough, by the end of the decade, to hold with high probability.
Starting roughly in 2006, the 10s, 100s, and 1000s became sinks (rows
three through five in figures 5.14). The probability of staying at these
levels year on year increases gradually. The probability of rising out of
that range falls. At levels immediately above that sink (rows one and
two in figures 5.15), the probability of falling into the sink rises at the
clear expense of the probability of staying put or rising. This trend
predates, and is uninterrupted by, the CME merger.
Neither transition matrix support the hypothesis that exchange
management is an important factor in lifecycle patterns, much less
the hypothesis that CME’s systems and network effects boost trading
volumes substantially relative to competing exchanges.
Recent trends in NYMEX lifecycles and the importance of electronic
trading
While they do not show a strong influence from the CME acquisition,
figures 5.14 and 5.15 may speak to the influence of electronic trading.
Pronounced lifecycle trends on the NYMEX seem to begin in 2006,
when the exchange abruptly switched from open-outcry to electronic
trading. These trends mirror the more general tendency across deriva-
tives markets over the last decade, with more flexible trading at low
volumes, more contracts moving up to modest volumes, and a small
decline in the probability of trading at high levels.
As mentioned above, it is difficult to separate out the effects of
electronic trading per se from those of the whole suite of new tools
that arrived with electronic trading, such as clear swaps platforms.
NYMEX, within its specialization in energy contracts threatened
by Enron, was a pioneer in cleared swaps transactions. In 2003,
it launched ClearPort, the platform now used for all of the CME’s
cleared swaps trades. ClearPort was marketed as an electronic trading
system because it disseminated information about specialized swaps
trades via screens15. However, most cleared swaps transactions are 15 Reuters News. NYMEX launchesCleaport electronic trading system.,January 17 2003
negotiated bilaterally, over the phone. That means that ClearPort
trades are supported by electronic infrastructure, but they are not
fully electronic. So, while NYMEX’s abrupt shift from pit-based to
electronic trading offers a prime opportunity to isolate the influence of
electronic trading on derivatives volumes, the advent of cleared swaps
complicates both the analysis of NYMEX data and the definition of
electronic trading.
the lifecycle of derivatives contracts 171
To the extent that we can identify the influence of fully electronic
trading on its own, then 2006, the year that NYMEX abruptly closed
pit trading, should produce discontinuities in ongoing lifecycle trends.
2006 does indeed show evidence of a discontinuity. That evidence is
not overwhelming, but it does support the hypothesis that the large
changes in derivatives markets in the 2000s were driven specifically by
the switch to electronic trading.
Derivatives volumes by product type
Figure 5.16 shows expected value estimates for trading at year ten for
each product type. Derivatives based on US treasuries, and, to a lesser
extent, derivatives based on natural gas and stock indexes enjoy higher
expected volumes than other product types.
The distinction between these strong performers and most of the
other product types in the sample is appreciable but does not hold
with high probability. The 95 percent probability interval for each of
those high expected volume product types sits within the upper tails
of the distributions for other product types. The long upper tails that
shadow the three top performers are largely a function of uncertainty
in estimating the parameter for relatively uncommon product types
rather than stellar historical performance. They reflect the fact that
we have relatively few observations of derivatives based on wood prod-
ucts, for example, and so our model allows for the possibility that
out-of-sample wood products may show high trading volumes in the
future.
Major currencies, grains, precious metals, petroleum-related prod-
ucts, and interest rates not derived from US treasuries define the mid-
dle of the pack for expected year ten volumes. They are joined by a
large group of product types whose expected volumes are subject to
great uncertainty, thanks to a scarcity of data.
Among these average performers, plastics and chemicals may be
promising niches for innovation. While their estimated expected vol-
umes are subject to considerable uncertainty, the data points we have
indicate that they are relatively strong performers.
On the low end of our expected year ten volume estimates are
single-stock futures16 and weather derivatives. Both are relatively new 16 This is consistent with figure 5.13
which shows OneChicago, the ex-
change specializing in single-stock
futures as a relative under-performer
in expected trading volume at year
ten.
product types with many correlated contracts launched in recent years.
Interestingly, these contract types appear to under-perform relative to
some product types like yield insurance and emissions in which trading
was effectively smothered by external events (the proliferation of sub-
sidized crop insurance in the US in the case of yield insurance and the
failure of the US to consistently promote cap-and-trade legislation in
the case of emissions).
172 direct climate markets
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
BASE METALS
CHEMICALS
CURRENCY
CURRENCY(NON−MAJOR)
DAIRY PRODUCTS
ELECTRICITY AND SOURCES
EMISSIONS
FIBER
FOODSTUFFS/SOFTS
Fertilizer
GRAINS
Interest Rates − U.S. Treasury
Interest Rates − non U.S. Treasury
LIVESTOCK/MEAT PRODUCTS
NARROW BASED INDICES
NATURAL GAS AND PRODUCTS
OILSEED and PRODUCTS
OTHER AGRICULTURAL
OTHER FINANCIAL INSTRUMENTS
PETROLEUM AND PRODUCTS
PLASTICS
PRECIOUS METALS
REAL ESTATE
SINGLE STOCK FUTURES
STOCK INDEX
WEATHER
WOOD PRODUCTS
Yield Insurance
1e+05 1e+07Expected volume after 10 years
Subg
roup
Figure 5.16: Expected trading volumeover ten years by product type
the lifecycle of derivatives contracts 173
vol 1's, t+1
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
0.18
0
0.06
0.08
0.03
0
0
0
0
0
0
0.25
0.02
0
0.03
0.06
0.03
0
0.07
0.19
0.25
0.06
0.25
0
0.16
0
0.01
0
BASE METALS
CHEMICALS
CURRENCY
CURRENCY(NON−MAJOR)
DAIRY PRODUCTS
ELECTRICITY AND SOURCES
EMISSIONS
FIBER
FOODSTUFFS/SOFTS
Fertilizer
GRAINS
INT RATES − NON U.S. TRES
INT RATES − U.S. TRES
LIVESTOCK/MEAT PRODUCTS
NARROW BASED INDICES
NATURAL GAS AND PRODUCTS
OILSEED and PRODUCTS
OTHER AGRICULTURAL
OTHER FINANCIAL INSTRUMENTS
PETROLEUM AND PRODUCTS
PLASTICS
PRECIOUS METALS
REAL ESTATE
SINGLE STOCK FUTURES
STOCK INDEX
WEATHER
WOOD PRODUCTS
Yield Insurance
vol 0, year t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Prod
uct s
ubgr
oup
vol 10's, t+1
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
0.18
0.01
0.22
0.03
0
0.13
0.11
0.6
0
0.19
0
0
0.14
0
0.05
0
0.22
0.11
0.05
0.06
0.22
0.06
0.44
0
0.11
0
0
0
BASE METALS
CHEMICALS
CURRENCY
CURRENCY(NON−MAJOR)
DAIRY PRODUCTS
ELECTRICITY AND SOURCES
EMISSIONS
FIBER
FOODSTUFFS/SOFTS
Fertilizer
GRAINS
INT RATES − NON U.S. TRES
INT RATES − U.S. TRES
LIVESTOCK/MEAT PRODUCTS
NARROW BASED INDICES
NATURAL GAS AND PRODUCTS
OILSEED and PRODUCTS
OTHER AGRICULTURAL
OTHER FINANCIAL INSTRUMENTS
PETROLEUM AND PRODUCTS
PLASTICS
PRECIOUS METALS
REAL ESTATE
SINGLE STOCK FUTURES
STOCK INDEX
WEATHER
WOOD PRODUCTS
Yield Insurance
vol 1's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Prod
uct s
ubgr
oup
Figure 5.17: Probability of transitionfrom annual volume in the hundredsto annual volume in the thousands(left) and from annual volume in thethousands to annual volume tens ofthousands (right) by product type
174 direct climate markets
vol 100's, t+1
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
0.17
0
0.25
0.05
0
0.32
0.13
0.18
0.21
0.2
0.08
0.04
0.36
0
0.17
0.32
0.17
0.18
0.04
0.25
0.17
0.13
0.03
0
0.04
0
0.59
0.06
BASE METALS
CHEMICALS
CURRENCY
CURRENCY(NON−MAJOR)
DAIRY PRODUCTS
ELECTRICITY AND SOURCES
EMISSIONS
FIBER
FOODSTUFFS/SOFTS
Fertilizer
GRAINS
INT RATES − NON U.S. TRES
INT RATES − U.S. TRES
LIVESTOCK/MEAT PRODUCTS
NARROW BASED INDICES
NATURAL GAS AND PRODUCTS
OILSEED and PRODUCTS
OTHER AGRICULTURAL
OTHER FINANCIAL INSTRUMENTS
PETROLEUM AND PRODUCTS
PLASTICS
PRECIOUS METALS
REAL ESTATE
SINGLE STOCK FUTURES
STOCK INDEX
WEATHER
WOOD PRODUCTS
Yield Insurance
vol 10's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Prod
uct s
ubgr
oup
vol 1000's, t+1
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
!
0.16
0.76
0.23
0.1
0.08
0.28
0.07
0.2
0.22
0.2
0.12
0.1
0.19
0
0.19
0.34
0.05
0.03
0.17
0.2
0.18
0.16
0
0.04
0.1
0.26
0
0
BASE METALS
CHEMICALS
CURRENCY
CURRENCY(NON−MAJOR)
DAIRY PRODUCTS
ELECTRICITY AND SOURCES
EMISSIONS
FIBER
FOODSTUFFS/SOFTS
Fertilizer
GRAINS
INT RATES − NON U.S. TRES
INT RATES − U.S. TRES
LIVESTOCK/MEAT PRODUCTS
NARROW BASED INDICES
NATURAL GAS AND PRODUCTS
OILSEED and PRODUCTS
OTHER AGRICULTURAL
OTHER FINANCIAL INSTRUMENTS
PETROLEUM AND PRODUCTS
PLASTICS
PRECIOUS METALS
REAL ESTATE
SINGLE STOCK FUTURES
STOCK INDEX
WEATHER
WOOD PRODUCTS
Yield Insurance
vol 100's, t
0.00 0.25 0.50 0.75 1.00Pr(year−on−year move)
Prod
uct s
ubgr
oup
Figure 5.18: Probability of transitionfrom annual volume in the hundredsto annual volume in the thousands(left) and from annual volume in thethousands to annual volume tens ofthousands (right) by product type
the lifecycle of derivatives contracts 175
Interestingly, single-stock futures and weather derivatives were more
likely than most other product types to climb up from low levels of
trading volume. Figures 5.17 and 5.18 present the probability of any
contract moving to higher levels of annual trading volume for each
product type in the sample. Single-stock futures are particularly likely
to recover from years of zero trading volume and weather derivatives
are particularly likely to move up to annual trading volumes in the
thousands. (See Lifecycle Appendix’s figures 13 and 14 for additional
details.) Insofar as these product types move fluidly up and down from
low levels of annual trading volumes they are representative of recent
trends across derivatives markets.
Conclusions
In this article we have presented a comprehensive analysis of trading
volumes for derivatives reported to exchanges in the United States.
Looking across decades, exchanges, and product types we see multi-
ple trends that challenge or significantly modify findings of existing
studies.
While a larger percentage of contracts had little or no volume in
any given year of the 2000s, contracts did not fail at the high rates
noted in previous analyses. Instead, they remained at low levels of
trading until they were needed, transitioning back into active trading
with greater probability than in previous decades. Interestingly, this
flexibility from low levels of trading meant that the long term outlook
for a new contract did not erode despite remarkable levels of new
contract innovation. During the 2000s the expected volume of a new
contract after ten years was above that of the 1990s and within the
range of previous decades. On balance, the explosion of innovation
catalyzed by electronic trading did not hurt the prospects for the
marginal contract.
We find that expected year ten trading volumes varied more decade
to decade than from exchange to exchange or product type to product
type. In particular, the lifecycle of a derivative on any given exchange
was largely indistinguishable from that on any other, with the likely
exception of OneChicago, which specializes in single-stock futures.
We find evidence that the decadal changes in derivative lifecycles
were driven by the switch to electronic trading rather than the consol-
idation of exchanges by looking at trends on the New York Mercantile
Exchange. The effects of electronic trading are difficult to separate
from the the related innovation of cleared swaps. However, trends in
NYMEX volumes following the 2006 launch of widespread electronic
trading tentatively support the hypothesis that electronic trading is
indeed driving recent trends in derivatives lifecycles across all sampled
176 direct climate markets
markets.
The statistical characteristics of derivative volumes
In addition to facilitating quick distributional comparisons across var-
ious contract groupings (decade, exchange, and product type), our
framework (Markov models) allows us to explore some basic ques-
tions about derivative markets in general. For example, based on our
Markov models it appears that trading volumes do not follow a normal
or log-normal random walk over time. In figures 5.4 and 5.5 it is clear
that the probability of remaining at a given level of annual volume
varies dramatically from one level to the next. These differences hold
with greater than 95 percent probability as do variations in the volume
dynamics across time (indicating that normal or log-normal models
of trading volume would suffer from stationarity problems as well).
Furthermore, switches to higher and lower levels of trading are often
not symmetric. In particular, an outright crash to zero trading vol-
ume appears more likely than would be predicted by a symmetrically
distributed random walk.
However, our analysis does affirm the common observation that
it is unusual for a contract to experience initial popularity and to
crash subsequently17. After reaching a trading volume in the tens of 17 E.T. Johnston and J.J. McConnell.Requiem for a market: An analysis ofthe rise and fall of a financial futurescontract. Review of Financial Studies,2(1):1–23, 1989
thousands, the probability that a contract will have annual trading
volume of zero in the subsequent year drops appreciably.
Optimal contract innovation
Much of the literature on derivative innovation focuses on the problem
of choosing the optimal derivatives contract to launch next. This
analysis does not directly address that question, but it does present
some trends relevant to previous theoretical work which could inform
further investigation.
One interpretation of Duffie and Jackson [1989] provides that rev-
enue maximizing marginal innovations are uncorrelated with existing
contracts. However, recent trends suggest that one of the key assump-
tions underlying this finding only holds weakly. Historically, correlated
contract innovations have not shown diminishing marginal volumes.
In general, innovation in derivatives markets has exploded in the last
decade seemingly without dragging down expected trading volumes
at year ten. Indeed, some of the highest volume product types (in ex-
pected volume terms) are highly correlated both to other derivatives of
their product category but also to the average returns of the economy
as a whole (US treasuries and stock indexes).
Tashjian and Weissman [1995] explains the proliferation of corre-
lated (and often redundant) contracts as a form of price discrimina-
the lifecycle of derivatives contracts 177
tion. They assume, that an exchange can charge higher fees on the
transaction for parties with larger and more concentrated exposure to
a given underlying index. This framework for understanding product
innovation holds up well in light of recent trends. As we have dis-
cussed, many recently launched contracts are cleared swaps, which
tend to be more specialized than conventional futures or options. As
Tashjian and Weissman [1995] predicted, exchanges charge a sub-
stantial premium on these specialized transactions. Fees on CME’s
ClearPort platform were more than 350 percent those on conven-
tional electronic futures trades as of late 201218. However, Tashjian 18 CME Group. Press release: CME
Group volume averaged 11.9 mil-
lion contracts per day in septem-
ber 2012, up 16 percent from au-
gust 2012, October 2012. URL
http://investor.cmegroup.com/
investor-relations/releasedetail.
cfm?ReleaseID=710622. Thanks to
Silla Brush of Bloomberg for directing
us to this press release
and Weissman [1995] suggested that exchanges would charge more for
single-asset-based derivatives (such as gold) than for derivatives that
represented the holding of multi-product firms (like the crack-spreads
used to reproduce petroleum refiners’ returns). In practice, cleared-
swaps contracts, with their relatively high fees, appear to be biased
toward the latter.
A third explanation of recent patterns in derivative innovation is
psychological. Based on Tversky et al. [1981], Shiller [1994] suggests
that a hedge “appears more attractive when it’s presented as the elim-
ination of risk rather than when it is described as the reduction of
risk.” This tendency to overvalue hedges tailored to the needs of spe-
cific firms may explain the proliferation of correlated contracts (and
their relative success) above and beyond the price discrimination sug-
gested in Tashjian and Weissman [1995].
What kind of economic goods are derivatives?
The present analysis could also suggest new ways of understanding
the economic value of derivatives. If indeed derivatives are simply
contingent contracts that move cash flows across time and states of
nature, then they should derive all their value from the way that they
mesh with hedger’s risk preferences. It follows from that idea, that
if risk preferences remain stable over time, then derivative trading
patterns should also remain stable.
But trading patterns have not been stable over the last decade.
Instead, they bear a striking qualitative resemblance to those of infor-
mation goods, particularly media:
• Each class of economic goods was, until recently, simple to classify:
normal goods with elastic demand and network effects.
• Starting a new derivatives market, just like producing a new mu-
sic album or launching a magazine, was a high risk, high reward
proposition.
• In the last decade both saw paradigm shifts in their marginal cost
178 direct climate markets
structure (i.e. there were fundamental changes in supply).
• At the same time, new technologies allowed consumers ubiquitous
access to goods (i.e. there were also fundamental changes in de-
mand).
• After those twin revolutions, markets:
- Rewarded specialty products more than in the past;
- Hosted blockbusters as large as/larger than ever;
- Did not offer the same opportunities for strong but less-than-
blockbuster products.
In media (and informational goods more generally) this transition
has upended many long-profitable business models and catalyzed a
great deal of innovation. In derivatives it has certainly opened up the
door to many new entrants like ICE, which now is one of two large
futures exchanges in the US today. But it is not clear whether those
new entrants are using fundamentally new business models.
How strong is the parallel? Should economists study derivatives
alongside informational goods? What does the possible connection
suggest for the future of derivatives? We believe that these questions
provide a solid foundation for future research.
